This RO1 response to RFA DA-99-003 proposes a multi-site whole- genome search for chromosomal loci influencing early-onset antisocial drug dependence. This project extends Colorado's ongoing sib-pair study of clinical probands with severe SD and CD, based on the QTL methodology developed by our colleagues, Fulker, Cherny et al. These disorders have significant heritability. Ascertainment, assessment, and collection of blood from probands and family members will be conducted through clinical research programs in three cities. With some 900 families and 3200 individuals there will be sufficient power to detect quantitative trait loci (QTL) of modest individual effect, while permitting a test of the robustness of QTL detection across independent sites. Our molecular genetic protocol will include transformation, establishment and maintenance of cell lines, and genotyping at our molecular facility. Data management, analysis, and data sharing will be conducted through our existing bioinformatics group using Discovery Manager\TM data management software and QTL analysis programs developed by us for quantitative measures in selected samples. This highly selected sample of probands and their siblings are assessed using standardized structured psychiatric interviews for DSM-IV diagnoses and symptom counts. QTL analyses will use symptom counts as a quantitative phenotype providing maximum information from affected and unaffected individuals. The specific aims of the current proposal are: (1) To identify several chromosomal regions, each less than 20 centiMorgans (cM) in length, which are likely to contain one or more genes influencing the development of early-onset dependence on drugs, including alcohol. (2) To identify several chromosomal regions which are very likely to contain one or more genes influencing the development of adolescent Conduct Disorder. (3) To determine whether the chromosomal regions associated with drug dependence overlap those associated with Conduct Disorder; this would imply that tightly- linked or identical genes underlie the frequent association of early-onset drug dependence and Conduct Disorder. (4) To provide adequate power for all of these analyses through a multi-site collaboration, resulting in an expanded sample carefully phenotyped by outstanding clinical researchers in adolescent substance dependence. (5) To establish, for the use of qualified researchers, a data library, which will include posted genotype and phenotype information, together with lymphoblastoid cell lines, from subjects of this research.